Intelligent power grid abnormal power consumption behavior detection method

Abstract

The invention discloses an intelligent power grid abnormal power consumption behavior detection method. The method comprises the following steps: power consumption behaviors of users are described through change of different power consumption service types of electric energy consumption amounts of the users along with time so that power consumption features of the users are fully reflected; the users are clustered by use of similarities between the power consumption behaviors of the users, deviations of abnormal power consumption behaviors from normal behavior modes are depicted by use of mutual association between the power consumption behaviors of the users in the same user cluster and correlation between the power consumption behaviors of the users and historical power consumption behaviors, such that effectiveness of abnormal power consumption behavior detection is guaranteed; through analyzing accumulated user power consumption behavior data in a fixed time interval, under the condition that the detection precision requirement is satisfied, too much noise is prevented from being introduced in analysis of vast real-time data, and the method is enabled to have quite good dynamical adaptability; and through analyzing the accumulated user power consumption behavior data in the fixed time interval, the original data amount is reduced, and the detection method disclosed by the invention can satisfy the requirement for real-time performance.

Description

technical field [0001] The invention relates to the field of power grids, in particular to a technology for detecting abnormal power consumption behaviors of smart grids. Background technique [0002] Smart grid power consumption behavior refers to the power usage behavior of users (such as residential users, commercial users, etc.) in the smart grid. Correspondingly, the abnormal power consumption behavior of the smart grid is manifested as the deviation or violation of the user's power consumption behavior and its normal behavior mode in the smart grid. [0003] Abnormal power consumption behavior in the smart grid may be caused by hackers attacking the system, such as tampering with the power usage records of smart meters; The purpose of electricity consumption is changed; or it is caused by the abnormal state of electrical equipment, such as the increase of electricity consumption due to failure of some equipment or aging of lines. These abnormal power consumption beha...

AI Technical Summary

Problems solved by technology

[0005] 1. In the smart grid, attackers may attack anywhere in the system and tamper with relevant data; well-designed attacks cannot be observed in state estimation based on current, voltage, phase, etc., nor can they be detected by any bad Data recognizer detects

[0006] 2. The smart grid can obtain more and more detailed parameters than the traditional power grid through advanced measurement, sensing, communication and other technologies, so as to describe the real-time power consumption behavior of each user in detail. However, all users' The real-time power consumption data is extremely large, and it is challenging to ensure the real-time detection of abnormal power consumption behaviors

This type of method can detect the deviation of power consumption behavior compared with its historical behavior or abnormal power consumption behavior caused by equipment failure lights, but it is difficult to detect well-designed false data injection attacks (FalseDataInjection Attack, FDIA)

Moreover, the modeling and analysis calculation complexity of this type of method is high, and it is difficult to guarantee the real-time detection of abnormal power consumption behavior in practice.

Images